A raging debate on whether artificial intelligence (AI) would eventually overtake human capabilities notwithstanding, a team of researchers has developed a system to provide superhuman vision to robots. The team, from the University of Pennsylvania School of Engineering and Applied Science (Penn Engineering), has developed ‘PanoRadar’, a new tool that transforms simple radio waves into detailed, 3D views of the environment.
Providing better vision to robots has always been a big challenge for them to operate in poor weather conditions. Traditional, light-based vision sensors like cameras or Light Detection And Ranging (LiDAR) have been found to fail in foggy conditions or heavy smoke. But radio waves, whose wavelengths are orders of magnitude longer than light waves, are found to penetrate smoke and fog to see through certain materials — something human beings cannot. PanoRadar, a sensor that operates like a lighthouse, sweeps its beam in a circle to scan the surroundings, using a system with a rotating vertical array of antennas.
As they rotate, the antennas send out radio waves and listen for their reflections from the environment, much like how a lighthouse’s beam reveals the presence of ships and coastal features. The researchers presented their work at the 2024 International Conference on Mobile Computing and Networking (MobiCom), where they described how PanoRadar leverages radio waves and AI to let robots navigate even the most challenging environments, like smoke-filled buildings or foggy roads.
AI helps PanoRadar go beyond the simple scanning strategy. It combines measurements from all rotation angles to enhance its imaging resolution. The rotation creates a dense array of virtual measurement points. This allows PanoRadar to achieve imaging resolution comparable to LiDAR. Signal processing and machine learning algorithms extract rich 3D information from the environment. It is set to make robots highly effective in service.
Science is a world in itself. Here are some interesting facts that connect with you
Nirad Mudur